基于修正残差的G-Markov模型的河南省小麦产量预测

Forecast of Henan wheat yield based on modified residual error of G-Markov model

粮食产量是一个非平稳的时间序列,结合经典灰色理论和Markov模型所产生的修正残差的G-Markov模型,能较好地对非稳定时间序列的状态转移行为进行预测.以河南省小麦产量预测为例,在传统GM(1,1)模型的基础上,对其预测值与实际值的残差序列进行Markov模型预测,更好地发挥了两个模型的优势.经检验证明,修正残差的G-Markov模型在小麦产量预测方面比传统的灰色GM(1,1)模型具有更高的精确度.

Food production is a non-stationary time series,combined with the classical grey theory and Markov model produced by the correction of residual G-Markov model,can be well state transition behavior of unsteady time series prediction.Based on the wheat yield prediction of Henan Province as an example,in the traditional GM(1,1)model,on the basis of the predicted values and actual values of Markov model to predict residual sequence,better play to the advantages of the two models.Proved by inspection and correction of residual G-Markov model on wheat yield prediction than the traditional grey GM(1,1)model has higher accuracy.